5D3 at one week: a few surprises

[victorwol]

You mean you don't get 6fps with that card after fill up the buffer? Because the 6fps should not depend in the fist set of photos on the memory card, all goes to the buffer in first place, so 6fps should work no matter which card you use. I tested with a very old card, taking burts photos of a digital chronographer and got exactly 6 fps. Even with a fast card I don't think you will get 6fps after filling the buffer with RAW... Unless of course you are talking about JPGs.....

 

[t.linn]

2: It's got something to do with the "real" aperture of the lens, not just the normalised one that the camera tells us. eg, at 100mm f/4.0, the aperture should be 25mm. But is it? 100/3.5=28.5mm, 100/4.5=22.2mm. Taking halfway in between all we can say is that the aperture is somewhere between 23.6mm and 26.8mm (at which point the aperture would bump to the next number up or down). Roughly. So some lenses that have "f/4.0" printed on the barrel that are closer to 26mm might work, but those that also have "f/4.0" but are in reality closer to 24mm might not work.

That certainly seems plausible, dr croubie. But do you think that explains why, with certain lenses, some of a certain type of cross point work while other cross points with the same spec do not work?

 

[t.linn]

the only way to switch to the real silent mode is through the AF drive button or the Q button +
be sure to check on the lcd display, that the shooting mode is followed by an 'S'

I find the resulting sound not only silent, but also incredibly sexier due to the small increase in its duration
Made me forget the increase in price.

Well I'm in favor of anything that will make me forget the price! But, seriously, I was using the drive mode button to switch between modes as you describe and I come to a different conclusion. I would not call the mode "silent"; I would call it "somewhat quieter". I'm glad to read that it is working so well for you.

 

[t.linn]

It has to do with the location of the exit pupil of the lens.

AF units have angular sensitivity. An f/2.8 focusing sensor has a spread of 20 degrees, meaning one line looks at one side of the exit pupil (the circle that is the virtual aperture, as you look at the back of the lens) and the other looks at the other side, and there's a 20 degree difference. For f/5.6 sensors, it's 10 degree.

Now with a lens such as a macro lens, the exit pupil can be much further up in the lens, as you focus closer and closer. So even if it's ostensibly an "f2.8" lens, since the exit pupil is so far away, it's outside the angle of view of the f/2.8 sensors.

If you look up "bellows factor," you'll see that at a magnification of 1:1, lenses lose two stops of aperture. So an f/2.8 lens is really f/5.6, at 1:1.

There are other lenses that lose AF points for similar reasons. The legendary 1200mm f/5.6L can only use the central cluster of AF points, for instance. As can be easily visualized, the exit pupil is probably pretty far up the lens barrel, making it invisible to the AF points that aren't near the center.

Really, it's all about geometry.

Thank you, Stephen. This makes perfect sense.

 

[t.linn]

You mean you don't get 6fps with that card after fill up the buffer? Because the 6fps should not depend in the fist set of photos on the memory card, all goes to the buffer in first place, so 6fps should work no matter which card you use. I tested with a very old card, taking burts photos of a digital chronographer and got exactly 6 fps. Even with a fast card I don't think you will get 6fps after filling the buffer with RAW... Unless of course you are talking about JPGs.....

Hmmm. I'll have to re-test, victorwol. I was not talking about after the buffer filled up. I was talking about right off the bat. This was one of the first things I tested after the unboxing. It was pretty obvious that I wasn't getting 6fps. Got home, switched to a newer card, got 6fps. You experience leads me to wonder if I was not using the "H" mode in my initial tests. I'll check again and update my results.

 

[t.linn]

t.linn NICE REPORT!... I have a question???? When using AEB...can you have exposures that are longer than 30sec. ...or do you have to put the camera on Bulb for any shutter speeds longer than 30 sec?, (like the 5D Mark II).

I do a LOT of HDR shooting and I have a Promote Unit (which is like hauling around a colostomy bag with my camera with 2 tubes, manual set-up. etc.. ....NOT a fluid way to work to say the least....but the only way I can shoot over 3 AEB shots without touching the camera... even with the Promote Unit I HAVE to touch the camera to get any exposures over 30secs.). The expanded AEB feature alone on the MkIII may get me to sell my II and purchase the upgrade...but being able to go longer than 30sec. in a bracket will be icing on the cake.
I don't understand why this is such a "slow-to-come-about" feature. There is a computer under the hood here...it should be pretty straight-forward to satisfy what I would call a basic photographic need. Timed long shutter releases. Right?

Canon DSLR cameras are limited to a maximum shutter speed of 30 sec, period. All shooting conditions, including AEB or HDR options have to live within that constraint. To automatically achieve exposures longer than 30 sec requires a fundamental camera design change - extending the 30 sec maximum to something larger. But what would you pick? Right now it's controlled by BULB!

Infared is correct. Still 30 second max. However, I agree with you, fotoray. I don't see why changing the 30 second limit would require a fundamental camera design change. (That's not to say that this isn't true.) Heck, I don't see why Canon couldn't build an intervalometer into the 5D3 itself. Haven't higher-end Nikon bodies been able to do this for awhile?

As for the colostomy bag Promote Unit, have you seen this? http://www.kickstarter.com/projects/1435018402/trigger-happy-camera-remote If you carry an Android or iPhone with you, this might just be the ticket. For only $55, you can lose the Promote Unit and add another tool to the Swiss Army Knife that our smartphones have become. (Having said that, I have not compared the features of the Trigger Happy to the Promote.)

 

[bp]

...However, I think it would be a mistake to conclude that 5D3 doesn't improve on image quality when compared to the 5D2. Better focused images improve image quality and better exposed images improve image quality...

I think this is the part that really hits home for me as well.

I've sat there with both 5D2 and 5D3, doing comparison shots using the same lens and getting perfect focus through live view, etc... etc... and yeah - the comparison between the two shots is shockingly similar. The MK2 can still produce amazing images if focus and exposure settings are perfect. My MK3 is ever-so-slightly sharper (some people say their MK2 is ever-so-slightly sharper - I think it's simple variance between copies). BUT out in real world situations, you don't have the luxury of taking your time and getting perfect focus with live view... My keeper rate in actual shooting situations has improved so dramatically, it's like I'm playing in a whole different game. Especially so with wide aperture lesnses